embedding-condensation

Measure layer-wise token embedding cosine similarity, assessing the severity of embedding condensation. Concept from [ICML 2026] Dispersion loss counteracts embedding condensation and improves generalization in small language models.

embedding-condensation has been downloaded 5,544 times in total on PyPI, including 233 in the last 30 days. The latest version is 2.0, released May 17, 2026.

Version2.0
Downloads
5.54k
LicenseNon-Commercial License — see ../LICENSE.md in the LM-Dispersion repository.
AuthorChen Liu
UpdatedMay 17, 2026

Downloads

Weekly, last 90d.
Includes CI traffic.

VersionsTotal2.*1.*
Range
View
Granularity
Group by
CI traffic
Stack: OffCI: Included3 / 5 series
Selected total3.04k
5.5kAll-time
233Last 30 days
11Last 24 h
<0.01/sPer second

Version distribution

Share of downloads by released version. Computed over the last quarter.

  • 01

    2.0

    766 downloads
    47.4%
  • 02

    1.1

    393 downloads
    24.3%
  • 03

    1.0.0

    264 downloads
    16.3%
  • 04

    0.1.0

    192 downloads
    11.9%

Guess the next day

Thirteen recent days of embedding-condensation downloads. Drag the green handle on the right to guess where day fourteen lands.

TRUTH1110SUNMONTUEWEDTHUFRISATSUNMONTUEWEDTHUFRI
    embedding-condensation · 5.5k downloads on PyPI